APPLICANT'S ABSTRACT: Pneumocystis Carinii pneumonia (PCP) is an opportunistic infection seen in many immunocompromised individuals, most notably in patients with acquired immunodeficiency syndrome (AIDS). Pentamidine is currently one of the drugs of choice in treating patients with AIDS. The drug, however, is associated with a high incidence of toxic side effects among which is hepatitis. Since alcohol consumption is a major risk factor for liver cirrhosis, alcoholic AIDS patients are at a greater risk of death from hepatotoxicity. There is, therefore, an urgent need for safer and less hepatotoxic drugs for treatment of PCP in this population of AIDS patients. At present, the molecular mechanism of action of pentamidine against Pneumocystis Carinii has not been firmly established. However, the drug is known to interact with a number of macromolecules including DNA in the pathogen. Being a flexible molecule, pentamidine can assume a number of interconvertible conformations. We hypothesize that this conformational flexibility allows pentamidine to bind to different macromolecules and this may account at least in part, for the therapeutic as well as toxic actions of the drug. It may therefore be possible to separate the therapeutic actions of pentamidine from its toxic actions via conformation-biological activity relationship studies. To test this hypothesis and achieve the goals of the project, we propose to conduct the following studies: a) design and synthesize conformationally restricted analogues related to pentamidine; b) evaluate the in vitro anti-PCP activity of the synthesized compounds in a P. carinii culture model; c) evaluate the in vivo anti-PCP activity and toxicity of the most promising compounds in an animal model of the disease; d) study the interactions of the proposed pentamidine analogues with DNA at the molecular level via thermal denaturation and molecular modeling studies. This section of the study is proposed because recent studies have indicated that the anti-PCP actions of pentamidine might be due to its interaction with the pathogenic genome. The information gained will be valuable not only in determining whether a correlation between drug- DNA interaction and anti-PCP efficacy or toxicity exist, but will also guide in the design of potentially more effective and safer anti-PCP agents.